CN109839624A - A kind of multilasered optical radar position calibration method and device - Google Patents
A kind of multilasered optical radar position calibration method and device Download PDFInfo
- Publication number
- CN109839624A CN109839624A CN201711204686.6A CN201711204686A CN109839624A CN 109839624 A CN109839624 A CN 109839624A CN 201711204686 A CN201711204686 A CN 201711204686A CN 109839624 A CN109839624 A CN 109839624A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- laser radar
- calibration object
- subset
- calibration
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 230000003287 optical effect Effects 0.000 title abstract 3
- 239000013598 vector Substances 0.000 claims abstract description 80
- 230000009466 transformation Effects 0.000 claims abstract description 41
- 239000011159 matrix material Substances 0.000 claims abstract description 31
- 238000013519 translation Methods 0.000 claims abstract description 30
- 230000008569 process Effects 0.000 claims abstract description 12
- 238000012216 screening Methods 0.000 claims description 36
- 238000010276 construction Methods 0.000 claims description 9
- 238000005070 sampling Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 5
- 230000000007 visual effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000002592 echocardiography Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
Landscapes
- Optical Radar Systems And Details Thereof (AREA)
Abstract
The present invention provides a kind of multilasered optical radar position calibration method and device, method includes: the first point cloud data of acquisition and the second point cloud data；First point cloud data and the second point cloud data are screened, first cloud subset and second point cloud subset are obtained；The first planar process vector n is estimated according to first cloud subset and second point cloud subset1With the second planar process vector n2, and construct the first equation: C*n1=n2；Transformation calibration object location, repeats m times, obtains the first equation of m group；Spin matrix C is obtained to the first equation solution of m group；Obtain the translation vector T between two radars；Using obtained spin matrix C and translation vector T as initial value, point cloud matching is carried out in first cloud subset of m group and second point cloud subset for collecting at m times, correspondence obtains m group C1And T1, by the highest C of score value in m group1And T1As final result.The present invention can accelerate the calibration process of multilasered optical radar positional relationship, while obtain the calibration result of degree of precision.
Description
Technical Field
The invention relates to the technical field of laser radars, in particular to a multi-laser-radar position calibration method and device.
Background
Laser radars are increasingly used because of their advantages such as high accuracy, large data volume, and high acquisition frequency. In the field of unmanned or assisted driving, lidar is an important sensor choice. The vehicle system uses the laser radar to construct an environment map, identify road information and avoid obstacles in real time. In the field of unmanned driving or auxiliary driving, one idea is to embed and arrange a plurality of multi-line laser radars around the vehicle body. In this scheme, the lidar is not typically scanned at full angle, but rather has a range of scanning (90 degrees or 180 degrees). The scanning fields of view between adjacent laser radars are overlapped to a certain extent. In the application scheme, the numerical relationship between the point cloud data acquired by different laser radars needs to be established, and the position relationship between the different laser radars needs to be accurate.
Fusion and matching of point cloud data of a plurality of laser radars are important technical subjects in the field of intelligent driving. One idea for multi-lidar position calibration is to find corresponding features between multiple lidar, such as coordinates of a target point between different lidar spatial coordinate systems. However, in general, it is difficult to ensure that different laser radars scan the same position of the target, especially when the point cloud on the target is sparse. The other idea is that a laser radar-camera combined calibration method of a visual sensor is utilized, namely the visual sensor is used for obtaining the accurate point coordinates of one point on a target, and the method has large workload and limited accuracy. Therefore, the simple, quick and high-precision multi-laser radar calibration method is called as an important technical problem.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a multi-laser-radar position calibration method and device, which can accelerate the calibration process of the multi-laser-radar position relation and obtain a calibration result with higher precision.
In order to achieve the purpose, the invention provides the following technical scheme:
a first aspect. The invention provides a multi-laser radar position calibration method, which comprises the following steps:
s1, placing the calibration object in the overlapping area of the first laser radar and the second laser radar, and collecting first point cloud data P formed by the first laser radar on the calibration object1And second point cloud data P formed by second laser radar points on the calibration object2；
S2, according to the first point cloud data P1Reflection intensity of different points and constraint condition of specified area on calibration object on first point cloud data P1Screening to determine a first point cloud subset P1 2(ii) a According to the second point cloud data P2The reflection intensity of different points and the constraint condition of the designated area on the calibration object to the second point cloud data P2Screening to determine a second point cloud subset P2 2；
S3, according to the first point cloud subset P1 2Estimating a first plane normal vector n of the calibration object in a first laser radar space coordinate system1According to the second point cloud subset P2 2Estimating a second plane normal vector n of the calibration object in a second laser radar space coordinate system2And according to the first plane normal vector n1And a second plane normal vector n2Constructing a first process: c n1＝n2(ii) a Wherein C represents a rotation transformation matrix between the first laser radar space coordinate system and the second laser radar space coordinate system;
s4, in the overlapping area of the first laser radar and the second laser radar, changing the position of the calibration object, and repeatedly executing the steps S1-S3 m times to obtain m groups of first equations;
s5, solving the m groups of first equations to obtain the rotation transformation matrix C;
s6, acquiring a translation vector T between the first laser radar and the second laser radar;
s7, transforming the rotation obtained in the step S5 into a matrixC and taking the translation vector T obtained in the step S6 as an initial value, respectively carrying out point cloud matching based on normal distribution transformation on m groups of first point cloud subsets and m groups of second point cloud subsets obtained by m-time collection and processing, and correspondingly obtaining m groups of rotation transformation matrixes C1And translation vector T1The rotation transformation matrix C with the highest value in m groups1And translation vector T1As a final result.
Further, when the position calibration of the N laser radars is needed, the steps S1-S7 are executed to circularly perform the position calibration of the two laser radars, and finally the position calibration of the N laser radars is completed, wherein N is greater than 2.
Further, the S2 specifically includes:
cloud data P of first point1Screening out points with middle reflection intensity larger than a first threshold value and located in a designated area on the calibration object as a first point cloud subset P1 2；
The second point cloud data P2Screening out points with middle reflection intensity larger than a second threshold and located in a designated area on the calibration object as a second point cloud subset P2 2；
Wherein the first threshold and the second threshold are the same or different.
Further, in the step S3, the cloud subset P is obtained according to the first point1 2Estimating a first plane normal vector n of the calibration object in a first laser radar space coordinate system1According to the second point cloud subset P2 2Estimating a second plane normal vector n of the calibration object in a second laser radar space coordinate system2The method specifically comprises the following steps:
for the first point cloud subset P1 2Carrying out plane estimation based on a random sampling consistency method to obtain a first plane normal vector n of the calibration object in a first laser radar space coordinate system1(ii) a For the second point cloud subset P2 2Performing random-based sampling consistencyThe plane estimation of the method is carried out to obtain a second plane normal vector n of the calibration object in a second laser radar space coordinate system2。
Further, the S6 specifically includes:
and acquiring a translation vector T between the first laser radar and the second laser radar in a measuring or estimating mode.
Further, m is more than or equal to 2.
Further, the calibration object is a flat plate calibration object.
Further, the designated area on the calibration object is a circular area on the flat plate calibration object.
Further, the designated area on the calibration object is a red or near-red circular area on the flat plate calibration object.
In a second aspect, the present invention further provides a multi-lidar position calibration apparatus, including:
a point cloud acquisition unit for placing a calibration object in the overlapping area of the first laser radar and the second laser radar, and acquiring first point cloud data P formed by the first laser radar on the calibration object1And second point cloud data P formed by second laser radar points on the calibration object2；
A point cloud screening unit for screening the first point cloud data P1Reflection intensity of different points and constraint condition of specified area on calibration object on first point cloud data P1Screening to determine a first point cloud subset P1 2(ii) a According to the second point cloud data P2The reflection intensity of different points and the constraint condition of the designated area on the calibration object to the second point cloud data P2Screening to determine a second point cloud subset P2 2；
An equation construction unit for constructing a set of point clouds P from the first point cloud subset1 2Estimating the position of the calibration object in a first lidarFirst plane normal vector n in space coordinate system1According to the second point cloud subset P2 2Estimating a second plane normal vector n of the calibration object in a second laser radar space coordinate system2And according to the first plane normal vector n1And a second plane normal vector n2Constructing a first process: c n1＝n2(ii) a Wherein C represents a rotation transformation matrix between the first laser radar space coordinate system and the second laser radar space coordinate system;
the system comprises a point cloud acquisition unit, a point cloud screening unit and an equation construction unit, wherein the point cloud acquisition unit is used for acquiring a point cloud of a target object, the point cloud screening unit is used for screening the point cloud of the target object, and the equation construction unit is used for constructing a first equation of m groups;
the solving unit is used for solving the m groups of first equations to obtain the rotation transformation matrix C;
the acquisition unit is used for acquiring a translation vector T between the first laser radar and the second laser radar;
an optimization unit, configured to perform, with the rotation transformation matrix C obtained by the solving unit and the translation vector T obtained by the obtaining unit as initial values, point cloud matching based on normal distribution transformation on m groups of first point cloud subsets and second point cloud subsets obtained by m-time processing by the point cloud screening unit, and correspondingly obtain m groups of rotation transformation matrices C1And translation vector T1The rotation transformation matrix C with the highest value in m groups1And translation vector T1As a final result.
According to the technical scheme, the multi-laser-radar position calibration method and the multi-laser-radar position calibration device provided by the invention have the advantages that firstly, the point cloud is quickly screened by utilizing the reflection intensity information of the laser radar, secondly, the planar normal vector is estimated by utilizing the screened point cloud, and then, the transformation relation between two laser radar coordinate systems is established by utilizing the estimated planar normal vector. Because the speed of screening the point cloud information based on the reflection intensity information is high, and the plane normal vector is a characteristic which is easy to obtain, the method can effectively accelerate the calibration process of the position relation of the multiple laser radars. In addition, on the basis of obtaining the initial value, the invention carries out more accurate point cloud matching to obtain an optimized calibration result, so that the invention can obtain a calibration result with higher accuracy.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a flowchart of a multi-lidar position calibration method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of the position relationship of multiple lidar systems with partially overlapping scan fields according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a flat panel calibration object applied to multi-lidar position calibration according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a point cloud collection according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a multi-lidar position calibration apparatus according to another embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
An embodiment of the present invention provides a method for calibrating multiple laser radar positions, referring to fig. 1, the method includes the following steps:
step 101: placing a calibration object in a superposed area of the fields of view of a first laser radar and a second laser radar, and collecting first point cloud data P formed by the first laser radar on the calibration object1And second point cloud data P formed by second laser radar points on the calibration object2。
In this step, see the schematic diagram of the positions and scanning ranges of two adjacent lidar shown in fig. 2. Wherein, the laser radar number on the left side is LiDAR1, and the laser radar number on the right side is LiDAR 2. A three-dimensional rectangular coordinate system is fixed on the laser radar, and the coordinate system is a right-hand coordinate system. The directions of the X axis and the Z axis are illustrated in fig. 2, the direction of the Y axis can be determined according to the rule of a right-hand coordinate system, and the scanning fields of the two laser radars in fig. 2 are overlapped to a certain extent. It should be noted that, the origin of the coordinate system selected by the laser radar is generally located inside the laser radar, and in fig. 2, for convenience of display, the origin of the coordinate system is specifically selected outside the laser radar. This arrangement is for convenience of illustration and does not change the technical principle of the present invention. Fig. 3 shows a calibration object of this embodiment, which is a flat plate with a red or near red circle on the flat plate. The size and the circular radius of the flat plate are comprehensively determined by the position relation of the two laser radars. The plate is denoted P in fig. 4.
In the step, the calibration object is placed in a field overlapping area of the first laser radar and the second laser radar, point cloud data of the two laser radars are collected respectively, and first point cloud data P are obtained1And second point cloud data P2。
It will be appreciated that the calibrant is preferably a plate calibrant. In addition, since the existing lidar generally uses near-infrared laser for spatial scanning, if the surface of the scanned object is red or near-red, the echo intensity is high, and therefore, in order to improve the accuracy of subsequent data calculation, the surface of the calibration object is preferably red or near-red.
Step 102: according to the first point cloud data P1Reflection intensity of different points and constraint condition of specified area on calibration object on first point cloud data P1Screening to determine a first point cloud subset P1 2(ii) a According to the second point cloud data P2The reflection intensity of different points and the constraint condition of the designated area on the calibration object to the second point cloud data P2Screening to determine a second point cloud subset P2 2。
It is understood that, at present, lidar applied to the field of unmanned and assisted driving can provide information representing the intensity of laser echoes in addition to three-dimensional spatial coordinate information. Therefore, the first point cloud data P in the above step 1011And second point cloud data P2The method not only comprises space coordinate information, but also comprises reflection intensity information.
It can be understood that the echo intensity indicates the reflection capability of the scanned object surface to the laser, and if the reflection capability is strong, the echo intensity is high, and if the reflection capability is weak, the echo intensity is low. Therefore, the reflection capability of the scanned object surface to the laser can be represented by the echo intensity value, which is denoted by the symbol r. When the surface of the scanned object is red or near red, the echo intensity is high, i.e. the value of the reflection intensity r is large.
As can be seen from the above analysis, in this step, the first point cloud data P may be extracted1Screening out points with middle reflection intensity larger than a first threshold value and located in a designated area on the calibration object as a first point cloud subset P1 2(ii) a Similarly, the second point cloud data P2The middle reflection intensity is greater than the second threshold value and is positioned on the targetScreening out points in a designated area on the fixed object as a second point cloud subset P2 2(ii) a Wherein the first threshold and the second threshold may be the same or different.
It is understood that the cloud data P may be at a first point by a first threshold value1Quickly selecting a point cloud subset P1 1The point cloud subset P1 1The spatial locations of most of the points are within a specified area of the flat plate landmark (e.g., within the circular area of the landmark as shown in fig. 4), but within the point cloud subset P1 1In the method, outliers may exist, so that the interference points can be effectively removed by combining the constraint condition of a designated area on the calibration object (such as the inside of a circular area shown in fig. 4). For example, if the point cloud subset P1 1Point-in-point distance point cloud subset P1 1Is greater than a certain value (optionally a radius of a circle on the flat calibration object), the point is removed, and finally the first cloud subset of points P is obtained1 2. Similarly, a second point cloud subset P can be obtained2 2。
Step 103: according to the first point cloud subset P1 2Estimating a first plane normal vector n of the calibration object in a first laser radar space coordinate system1According to the second point cloud subset P2 2Estimating a second plane normal vector n of the calibration object in a second laser radar space coordinate system2And according to the first plane normal vector n1And a second plane normal vector n2Constructing a first process: c n1＝n2(ii) a Where C represents a rotation transformation matrix between the first and second lidar spatial coordinate systems.
In this step, the first point cloud subset P is selected1 2Carrying out plane estimation based on a random sampling consistency method to obtain a first plane normal vector n of the calibration object in a first laser radar space coordinate system1(ii) a For the second point cloud subset P2 2Performing random-based sampling consensusPlane estimation of the linear method is carried out to obtain a second plane normal vector n of the calibration object in a second laser radar space coordinate system2。
In particular, according to said first point cloud subset P1 2Estimating the plane equation of the calibration object in the first laser radar space coordinate system by adopting a random sampling consistency method, and further obtaining a normal vector of the plane of the calibration object according to the plane equation of the calibration object, namely a first normal vector n of the plane1. Similarly, a similar method is adopted to obtain a second plane normal vector n2。
Obtaining a plane normal vector n of a calibration object under two coordinate systems1And n2Then, the following equation is constructed:
C*n1＝n2
and C is a third-order orthogonal matrix which represents the rotation transformation relation between the first laser radar space coordinate system and the second laser radar space coordinate system.
Step 104: and in the overlapping area of the fields of view of the first laser radar and the second laser radar, changing the position of the calibration object, and repeatedly executing the steps 101-103 m times to obtain m groups of first equations.
In this step, the value of m is generally at least 2, and the first equation for obtaining m groups is as follows:
wherein,representing that the first plane normal vector is obtained for the m-th time,and expressing that the second plane normal vector is obtained at the m-th time.
Step 105: and solving the m groups of first equations to obtain the rotation transformation matrix C.
In the step, a normal matrix N is constructed by using all plane normal vectors of the calibration object in the first laser radar space coordinate system1：
Similarly, a normal matrix N is constructed by using all plane normal vectors of the calibration object in the second laser radar space coordinate system2：
Thus, the above m sets of first equations are converted into the following matrix equations:
C·N1＝N2
further, the solution of the above matrix equation is:
step 106: and acquiring a translation vector T between the first laser radar and the second laser radar.
In this step, a translation vector T between the first lidar and the second lidar may be obtained by measurement or estimation.
Step 107: taking the rotation transformation matrix C obtained in the step 105 and the translation vector T obtained in the step 106 as initial values of a Normal Distribution Transformation (NDT) algorithm, respectively performing point cloud matching based on the Normal Distribution Transformation (NDT) on m groups of first point cloud subsets and second point cloud subsets obtained by m-time collection and processing, and correspondingly obtaining m groups of rotation transformation matrices C1And translation vector T1The rotation transformation matrix C with the highest NDT algorithm score in the m groups1And translation vector T1As a final result.
It will be appreciated that the rotation transformation matrix C with the highest score is obtained1And translation vector T1And then, the position calibration result between the first laser radar and the second laser radar can be obtained.
It should be noted that the order of the method steps shown in this embodiment is set merely to illustrate the technical principle of the present invention, and the method steps can be performed in parallel in actual operation.
As can be seen from the above description, in the multi-lidar position calibration method provided by this embodiment, first, point clouds are quickly screened by using lidar reflection intensity information, then, a plane normal vector is estimated by using the screened point clouds, and then, a transformation relationship between two lidar coordinate systems is established by using the estimated plane normal vector. Because the speed of screening point cloud information based on the reflection intensity information is high, and the plane normal vector is the characteristic which is easy to obtain, the calibration process of the position relation of the multiple laser radars can be effectively accelerated by the embodiment. In addition, on the basis of obtaining the initial value, the embodiment performs more accurate point cloud matching to obtain an optimized calibration result, so that the embodiment can obtain a calibration result with higher accuracy.
It can be understood that, when position calibration needs to be performed on N laser radars, the above steps 101 to 107 may be performed to perform position calibration on two laser radars in a cycle, and finally, position calibration on N laser radars is completed, where N is greater than 2.
Therefore, the spatial position relation between the multiple groups of laser radars can be accurately calibrated, and the operation is simple and quick.
Another embodiment of the present invention provides a multi-lidar position calibration apparatus, referring to fig. 5, the apparatus includes: the system comprises a point cloud acquisition unit 21, a point cloud screening unit 22, an equation construction unit 23, an equation system construction unit 24, a solving unit 25, an acquisition unit 26 and an optimization unit 27, wherein:
a point cloud collection unit 21, configured to place a calibration object in a region where the first and second lidar fields of view overlap, and collect first point cloud data P formed by the first lidar on the calibration object1And second point cloud data P formed by second laser radar points on the calibration object2；
A point cloud screening unit 22 for screening the first point cloud data P1Reflection intensity of different points and constraint condition of specified area on calibration object on first point cloud data P1Screening to determine a first point cloud subset P1 2(ii) a According to the second point cloud data P2The reflection intensity of different points and the constraint condition of the designated area on the calibration object to the second point cloud data P2Screening to determine a second point cloud subset P2 2；
An equation construction unit 23 for constructing a first point cloud subset P from the first point cloud subset1 2Estimating a first plane normal vector n of the calibration object in a first laser radar space coordinate system1According to the second point cloud subset P2 2Estimating a second plane normal vector n of the calibration object in a second laser radar space coordinate system2And according to the first plane normal vector n1And a second plane normal vector n2Constructing a first process: c n1＝n2(ii) a Wherein C represents a rotation transformation matrix between the first laser radar space coordinate system and the second laser radar space coordinate system;
an equation set constructing unit 24, configured to transform the position of the calibration object in a region where the fields of view of the first laser radar and the second laser radar overlap, and repeatedly execute the point cloud collecting unit 21, the point cloud screening unit 22, and the equation constructing unit 23 m times to construct m sets of first equations;
a solving unit 25, configured to solve the m groups of first equations to obtain the rotation transformation matrix C;
an obtaining unit 26, configured to obtain a translation vector T between the first laser radar and the second laser radar;
an optimizing unit 27, configured to perform, with the rotation transformation matrix C obtained by the solving unit 25 and the translation vector T obtained by the obtaining unit 26 as initial values, point cloud matching based on normal distribution transformation on m groups of first point cloud subsets and second point cloud subsets obtained by processing the point cloud screening unit 22m times, respectively, and correspondingly obtain m groups of rotation transformation matrices C1And translation vector T1The rotation transformation matrix C with the highest value in m groups1And translation vector T1As a final result.
The multi-lidar position calibration device provided by the embodiment can be used for executing the multi-lidar position calibration method provided by the embodiment, and the working principle and the technical effect are similar, so that the description is omitted here.
The above examples are only for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A multi-laser radar position calibration method is characterized by comprising the following steps:
s1, placing the calibration object in the overlapping area of the first laser radar and the second laser radar, and collecting first point cloud data P formed by the first laser radar on the calibration object1And second point cloud data P formed by second laser radar points on the calibration object2；
S2, according to the first point cloud data P1The reflection intensity of different points and the constraint condition of the designated area on the calibration object to the first point cloud numberAccording to P1Screening to determine a first point cloud subset P1 2(ii) a According to the second point cloud data P2The reflection intensity of different points and the constraint condition of the designated area on the calibration object to the second point cloud data P2Screening to determine a second point cloud subset P2 2；
S3, according to the first point cloud subset P1 2Estimating a first plane normal vector n of the calibration object in a first laser radar space coordinate system1According to the second point cloud subset P2 2Estimating a second plane normal vector n of the calibration object in a second laser radar space coordinate system2And according to the first plane normal vector n1And a second plane normal vector n2Constructing a first process: c n1＝n2(ii) a Wherein C represents a rotation transformation matrix between the first laser radar space coordinate system and the second laser radar space coordinate system;
s4, in the overlapping area of the first laser radar and the second laser radar, changing the position of the calibration object, and repeatedly executing the steps S1-S3 m times to obtain m groups of first equations;
s5, solving the m groups of first equations to obtain the rotation transformation matrix C;
s6, acquiring a translation vector T between the first laser radar and the second laser radar;
s7, respectively carrying out point cloud matching based on normal distribution transformation on m groups of first point cloud subsets and m groups of second point cloud subsets obtained by m-time collection and processing by taking the rotation transformation matrix C obtained in the step S5 and the translation vector T obtained in the step S6 as initial values, and correspondingly obtaining m groups of rotation transformation matrices C1And translation vector T1The rotation transformation matrix C with the highest value in m groups1And translation vector T1As a final result.
2. The method of claim 1, wherein when position calibration is required for N lidar, steps S1-S7 are performed to perform position calibration for two lidar, and finally N lidar position calibration is completed, where N > 2.
3. The method according to claim 1, wherein the S2 specifically includes:
cloud data P of first point1Screening out points with middle reflection intensity larger than a first threshold value and located in a designated area on the calibration object as a first point cloud subset P1 2；
The second point cloud data P2Screening out points with middle reflection intensity larger than a second threshold and located in a designated area on the calibration object as a second point cloud subset P2 2；
Wherein the first threshold and the second threshold are the same or different.
4. The method according to claim 1, wherein the first point cloud subset P is used in S31 2Estimating a first plane normal vector n of the calibration object in a first laser radar space coordinate system1According to the second point cloud subset P2 2Estimating a second plane normal vector n of the calibration object in a second laser radar space coordinate system2The method specifically comprises the following steps:
for the first point cloud subset P1 2Carrying out plane estimation based on a random sampling consistency method to obtain a first plane normal vector n of the calibration object in a first laser radar space coordinate system1(ii) a For the second point cloud subset P2 2Carrying out plane estimation based on a random sampling consistency method to obtain a second plane normal vector n of the calibration object in a second laser radar space coordinate system2。
5. The method according to claim 1, wherein the S6 specifically includes:
and acquiring a translation vector T between the first laser radar and the second laser radar in a measuring or estimating mode.
6. The method of claim 1, wherein m.gtoreq.2.
7. A method according to any one of claims 1 to 6, wherein the calibration object is a plate calibration object.
8. The method of claim 7, wherein the designated area on the target is a circular area on a flat plate target.
9. The method of claim 8, wherein the designated area on the target is a red or near red circular area on a flat plate target.
10. A multi-laser radar position calibration device is characterized by comprising:
a point cloud acquisition unit for placing a calibration object in the overlapping area of the first laser radar and the second laser radar, and acquiring first point cloud data P formed by the first laser radar on the calibration object1And second point cloud data P formed by second laser radar points on the calibration object2；
A point cloud screening unit for screening the first point cloud data P1Reflection intensity of different points and constraint condition of specified area on calibration object on first point cloud data P1Screening to determine a first point cloud subset P1 2(ii) a According to the second point cloud data P2The reflection intensity of different points and the constraint condition of the designated area on the calibration object to the second point cloud data P2Screening to determine a second point cloud subset P2 2；
An equation construction unit for constructing a set of point clouds P from the first point cloud subset1 2Estimating a first plane normal vector n of the calibration object in a first laser radar space coordinate system1According to the second point cloud subset P2 2Estimating a second plane normal vector n of the calibration object in a second laser radar space coordinate system2And according to the first plane normal vector n1And a second plane normal vector n2Constructing a first process: c n1＝n2(ii) a Wherein C represents a rotation transformation matrix between the first laser radar space coordinate system and the second laser radar space coordinate system;
the system comprises a point cloud acquisition unit, a point cloud screening unit and an equation construction unit, wherein the point cloud acquisition unit is used for acquiring a point cloud of a target object, the point cloud screening unit is used for screening the point cloud of the target object, and the equation construction unit is used for constructing a first equation of m groups;
the solving unit is used for solving the m groups of first equations to obtain the rotation transformation matrix C;
the acquisition unit is used for acquiring a translation vector T between the first laser radar and the second laser radar;
an optimization unit, configured to perform, with the rotation transformation matrix C obtained by the solving unit and the translation vector T obtained by the obtaining unit as initial values, point cloud matching based on normal distribution transformation on m groups of first point cloud subsets and second point cloud subsets obtained by m-time processing by the point cloud screening unit, and correspondingly obtain m groups of rotation transformation matrices C1And translation vector T1The rotation transformation matrix C with the highest value in m groups1And translation vector T1As a final result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711204686.6A CN109839624A (en) | 2017-11-27 | 2017-11-27 | A kind of multilasered optical radar position calibration method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711204686.6A CN109839624A (en) | 2017-11-27 | 2017-11-27 | A kind of multilasered optical radar position calibration method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109839624A true CN109839624A (en) | 2019-06-04 |
Family
ID=66880210
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711204686.6A Pending CN109839624A (en) | 2017-11-27 | 2017-11-27 | A kind of multilasered optical radar position calibration method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109839624A (en) |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110386147A (en) * | 2019-06-13 | 2019-10-29 | 北京百度网讯科技有限公司 | The calibration detection method and device of sensor applied to vehicle |
| CN110741282A (en) * | 2019-08-21 | 2020-01-31 | 深圳市速腾聚创科技有限公司 | External parameter calibration method and device, computing equipment and computer storage medium |
| CN110827339A (en) * | 2019-11-05 | 2020-02-21 | 北京深测科技有限公司 | Method for extracting target point cloud |
| CN111007485A (en) * | 2020-03-09 | 2020-04-14 | 中智行科技有限公司 | Image processing method and device and computer storage medium |
| CN111398936A (en) * | 2020-03-11 | 2020-07-10 | 山东大学 | Multi-path side laser radar point cloud registration device and using method thereof |
| CN111413689A (en) * | 2020-05-07 | 2020-07-14 | 沃行科技(南京)有限公司 | Efficient static calibration method for realizing multi-laser radar point cloud alignment based on rviz |
| CN111830498A (en) * | 2020-07-29 | 2020-10-27 | 成都纳雷科技有限公司 | Multi-radar automatic networking method and system based on millimeter wave region security |
| CN111965627A (en) * | 2020-08-18 | 2020-11-20 | 湖北亿咖通科技有限公司 | Multi-laser radar calibration method for vehicle |
| CN112485774A (en) * | 2020-11-26 | 2021-03-12 | 中国第一汽车股份有限公司 | Vehicle-mounted laser radar calibration method, device, equipment and storage medium |
| CN112509062A (en) * | 2020-12-17 | 2021-03-16 | 广东工业大学 | Calibration plate, calibration system and calibration method |
| CN112505704A (en) * | 2020-11-10 | 2021-03-16 | 北京埃福瑞科技有限公司 | Method for improving safety of train autonomous intelligent sensing system and train |
| WO2021051439A1 (en) * | 2019-09-26 | 2021-03-25 | 深圳市速腾聚创科技有限公司 | Calibration method, apparatus, storage medium and multi-channel lidar |
| CN112630751A (en) * | 2019-10-09 | 2021-04-09 | 中车时代电动汽车股份有限公司 | Calibration method of laser radar |
| CN112697057A (en) * | 2021-02-01 | 2021-04-23 | 南京耘瞳科技有限公司 | Method for detecting thickness of feeding belt |
| CN112835007A (en) * | 2021-01-07 | 2021-05-25 | 北京百度网讯科技有限公司 | Point cloud data conversion method and device, electronic equipment and storage medium |
| CN113156407A (en) * | 2021-02-24 | 2021-07-23 | 长沙行深智能科技有限公司 | Vehicle-mounted laser radar external parameter combined calibration method, system, medium and equipment |
| CN113495256A (en) * | 2020-04-01 | 2021-10-12 | 北京京东乾石科技有限公司 | Method and device for determining accuracy of calibration results among multiple laser radars |
| CN113534110A (en) * | 2021-06-24 | 2021-10-22 | 香港理工大学深圳研究院 | Static calibration method for multi-laser radar system |
| CN113759346A (en) * | 2020-10-10 | 2021-12-07 | 北京京东乾石科技有限公司 | Laser radar calibration method and device, electronic equipment and storage medium |
| WO2022001325A1 (en) * | 2020-06-30 | 2022-01-06 | 上海商汤智能科技有限公司 | Point cloud data fusion method and apparatus, electronic device, storage medium, and computer program |
| CN114076935A (en) * | 2020-08-20 | 2022-02-22 | 北京万集科技股份有限公司 | Laser radar and camera combined calibration method and device, server and computer readable storage medium |
| CN114755666A (en) * | 2022-06-01 | 2022-07-15 | 苏州一径科技有限公司 | Point cloud expansion evaluation method, device and equipment |
| WO2022179566A1 (en) * | 2021-02-26 | 2022-09-01 | 上海商汤智能科技有限公司 | Extrinsic parameter calibration method and apparatus, electronic device and storage medium |
| CN115236644A (en) * | 2022-07-26 | 2022-10-25 | 广州文远知行科技有限公司 | Laser radar external parameter calibration method, device, equipment and storage medium |
| CN117471433A (en) * | 2023-12-28 | 2024-01-30 | 广东威恒输变电工程有限公司 | Construction machinery laser point cloud real-time extraction method based on high reflection intensity target |
| CN117689536A (en) * | 2024-02-01 | 2024-03-12 | 浙江华是科技股份有限公司 | Laser radar splicing registration method, system, device and computer storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103837869A (en) * | 2014-02-26 | 2014-06-04 | 北京工业大学 | Vector-relation-based method for calibrating single-line laser radar and CCD camera |
| CN104677340A (en) * | 2013-11-30 | 2015-06-03 | 中国科学院沈阳自动化研究所 | Point character based monocular vision pose measurement method |
| CN106228537A (en) * | 2016-07-12 | 2016-12-14 | 北京理工大学 | A kind of three-dimensional laser radar and the combined calibrating method of monocular-camera |
| US9532031B1 (en) * | 2014-04-08 | 2016-12-27 | The United States Of America As Represented By The Secretary Of The Navy | Method for extrinsic camera calibration using a laser beam |
| US9574915B1 (en) * | 2015-12-22 | 2017-02-21 | National Chung-Shan Institute Of Science & Technology | Precision calibration method for high-precise rotary encoder |
| CN106556825A (en) * | 2015-09-29 | 2017-04-05 | 北京自动化控制设备研究所 | A kind of combined calibrating method of panoramic vision imaging system |
| CN106872963A (en) * | 2017-03-31 | 2017-06-20 | 厦门大学 | A kind of automatic Calibration algorithm of multigroup multi-line laser radar |
| CN107153186A (en) * | 2017-01-06 | 2017-09-12 | 深圳市速腾聚创科技有限公司 | Laser radar scaling method and laser radar |
| CN107220995A (en) * | 2017-04-21 | 2017-09-29 | 西安交通大学 | A kind of improved method of the quick point cloud registration algorithms of ICP based on ORB characteristics of image |
-
2017
- 2017-11-27 CN CN201711204686.6A patent/CN109839624A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104677340A (en) * | 2013-11-30 | 2015-06-03 | 中国科学院沈阳自动化研究所 | Point character based monocular vision pose measurement method |
| CN103837869A (en) * | 2014-02-26 | 2014-06-04 | 北京工业大学 | Vector-relation-based method for calibrating single-line laser radar and CCD camera |
| US9532031B1 (en) * | 2014-04-08 | 2016-12-27 | The United States Of America As Represented By The Secretary Of The Navy | Method for extrinsic camera calibration using a laser beam |
| CN106556825A (en) * | 2015-09-29 | 2017-04-05 | 北京自动化控制设备研究所 | A kind of combined calibrating method of panoramic vision imaging system |
| US9574915B1 (en) * | 2015-12-22 | 2017-02-21 | National Chung-Shan Institute Of Science & Technology | Precision calibration method for high-precise rotary encoder |
| CN106228537A (en) * | 2016-07-12 | 2016-12-14 | 北京理工大学 | A kind of three-dimensional laser radar and the combined calibrating method of monocular-camera |
| CN107153186A (en) * | 2017-01-06 | 2017-09-12 | 深圳市速腾聚创科技有限公司 | Laser radar scaling method and laser radar |
| CN106872963A (en) * | 2017-03-31 | 2017-06-20 | 厦门大学 | A kind of automatic Calibration algorithm of multigroup multi-line laser radar |
| CN107220995A (en) * | 2017-04-21 | 2017-09-29 | 西安交通大学 | A kind of improved method of the quick point cloud registration algorithms of ICP based on ORB characteristics of image |
Non-Patent Citations (6)
| Title |
|---|
| PETER BIBER: "The Normal Distributions Transform: A New Approach to Laser Scan Matching", 《CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS》 * |
| UNNIKRISHNAN RANJITH等: "Fast Extrinsic Calibration of a Laser Rangefinder to a Camera", 《ROBOTICS INSTITUTE，CARNEGIE MELLON UNIVERSITY》 * |
| 张晓等: "基于改进正态分布变换算法的点云配准", 《中国激光》 * |
| 杨飚等: "基于正态分布变换域迭代最近点的快速点云配准算法", 《科学技术与工程》 * |
| 王力: "基于人工标志的激光扫描数据自动拼接技术研究", 《中国优秀硕士学位论文全文数据库 基础科技辑》 * |
| 程金龙等: "车载激光雷达外参数的标定方法", 《光电工程》 * |
Cited By (42)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110386147B (en) * | 2019-06-13 | 2020-09-01 | 北京百度网讯科技有限公司 | Calibration detection method and device applied to sensor of vehicle |
| CN110386147A (en) * | 2019-06-13 | 2019-10-29 | 北京百度网讯科技有限公司 | The calibration detection method and device of sensor applied to vehicle |
| CN110741282A (en) * | 2019-08-21 | 2020-01-31 | 深圳市速腾聚创科技有限公司 | External parameter calibration method and device, computing equipment and computer storage medium |
| CN112955778A (en) * | 2019-09-26 | 2021-06-11 | 深圳市速腾聚创科技有限公司 | Correction method, correction device, storage medium and multi-channel laser radar |
| WO2021051439A1 (en) * | 2019-09-26 | 2021-03-25 | 深圳市速腾聚创科技有限公司 | Calibration method, apparatus, storage medium and multi-channel lidar |
| CN112955778B (en) * | 2019-09-26 | 2023-08-04 | 深圳市速腾聚创科技有限公司 | Correction method, correction device, storage medium and multichannel laser radar |
| CN112630751A (en) * | 2019-10-09 | 2021-04-09 | 中车时代电动汽车股份有限公司 | Calibration method of laser radar |
| CN110827339B (en) * | 2019-11-05 | 2022-08-26 | 北京深测科技有限公司 | Method for extracting target point cloud |
| CN110827339A (en) * | 2019-11-05 | 2020-02-21 | 北京深测科技有限公司 | Method for extracting target point cloud |
| CN111007485A (en) * | 2020-03-09 | 2020-04-14 | 中智行科技有限公司 | Image processing method and device and computer storage medium |
| CN111398936A (en) * | 2020-03-11 | 2020-07-10 | 山东大学 | Multi-path side laser radar point cloud registration device and using method thereof |
| CN111398936B (en) * | 2020-03-11 | 2021-04-06 | 山东大学 | Multi-path side laser radar point cloud registration device and using method thereof |
| CN113495256B (en) * | 2020-04-01 | 2024-04-05 | 北京京东乾石科技有限公司 | Method and device for determining accuracy of calibration result among multiple laser radars |
| CN113495256A (en) * | 2020-04-01 | 2021-10-12 | 北京京东乾石科技有限公司 | Method and device for determining accuracy of calibration results among multiple laser radars |
| CN111413689B (en) * | 2020-05-07 | 2023-04-07 | 沃行科技(南京)有限公司 | Efficient static calibration method for realizing multi-laser radar point cloud alignment based on rviz |
| CN111413689A (en) * | 2020-05-07 | 2020-07-14 | 沃行科技(南京)有限公司 | Efficient static calibration method for realizing multi-laser radar point cloud alignment based on rviz |
| JP2022541976A (en) * | 2020-06-30 | 2022-09-29 | シャンハイ センスタイム インテリジェント テクノロジー カンパニー リミテッド | Point cloud data fusion method, device, electronic device, storage medium and computer program |
| WO2022001325A1 (en) * | 2020-06-30 | 2022-01-06 | 上海商汤智能科技有限公司 | Point cloud data fusion method and apparatus, electronic device, storage medium, and computer program |
| CN111830498A (en) * | 2020-07-29 | 2020-10-27 | 成都纳雷科技有限公司 | Multi-radar automatic networking method and system based on millimeter wave region security |
| CN111965627A (en) * | 2020-08-18 | 2020-11-20 | 湖北亿咖通科技有限公司 | Multi-laser radar calibration method for vehicle |
| CN114076935A (en) * | 2020-08-20 | 2022-02-22 | 北京万集科技股份有限公司 | Laser radar and camera combined calibration method and device, server and computer readable storage medium |
| CN113759346A (en) * | 2020-10-10 | 2021-12-07 | 北京京东乾石科技有限公司 | Laser radar calibration method and device, electronic equipment and storage medium |
| CN112505704B (en) * | 2020-11-10 | 2024-06-07 | 北京埃福瑞科技有限公司 | Method for improving safety of autonomous intelligent perception system of train and train |
| CN112505704A (en) * | 2020-11-10 | 2021-03-16 | 北京埃福瑞科技有限公司 | Method for improving safety of train autonomous intelligent sensing system and train |
| CN112485774B (en) * | 2020-11-26 | 2024-03-15 | 中国第一汽车股份有限公司 | Vehicle-mounted laser radar calibration method, device, equipment and storage medium |
| CN112485774A (en) * | 2020-11-26 | 2021-03-12 | 中国第一汽车股份有限公司 | Vehicle-mounted laser radar calibration method, device, equipment and storage medium |
| CN112509062B (en) * | 2020-12-17 | 2023-09-12 | 广东工业大学 | Calibration plate, calibration system and calibration method |
| CN112509062A (en) * | 2020-12-17 | 2021-03-16 | 广东工业大学 | Calibration plate, calibration system and calibration method |
| CN112835007B (en) * | 2021-01-07 | 2023-04-18 | 北京百度网讯科技有限公司 | Point cloud data conversion method and device, electronic equipment and storage medium |
| CN112835007A (en) * | 2021-01-07 | 2021-05-25 | 北京百度网讯科技有限公司 | Point cloud data conversion method and device, electronic equipment and storage medium |
| CN112697057A (en) * | 2021-02-01 | 2021-04-23 | 南京耘瞳科技有限公司 | Method for detecting thickness of feeding belt |
| CN113156407B (en) * | 2021-02-24 | 2023-09-05 | 长沙行深智能科技有限公司 | Vehicle-mounted laser radar external parameter joint calibration method, system, medium and device |
| CN113156407A (en) * | 2021-02-24 | 2021-07-23 | 长沙行深智能科技有限公司 | Vehicle-mounted laser radar external parameter combined calibration method, system, medium and equipment |
| WO2022179566A1 (en) * | 2021-02-26 | 2022-09-01 | 上海商汤智能科技有限公司 | Extrinsic parameter calibration method and apparatus, electronic device and storage medium |
| CN113534110A (en) * | 2021-06-24 | 2021-10-22 | 香港理工大学深圳研究院 | Static calibration method for multi-laser radar system |
| CN113534110B (en) * | 2021-06-24 | 2023-11-24 | 香港理工大学深圳研究院 | Static calibration method for multi-laser radar system |
| CN114755666A (en) * | 2022-06-01 | 2022-07-15 | 苏州一径科技有限公司 | Point cloud expansion evaluation method, device and equipment |
| CN115236644A (en) * | 2022-07-26 | 2022-10-25 | 广州文远知行科技有限公司 | Laser radar external parameter calibration method, device, equipment and storage medium |
| CN117471433A (en) * | 2023-12-28 | 2024-01-30 | 广东威恒输变电工程有限公司 | Construction machinery laser point cloud real-time extraction method based on high reflection intensity target |
| CN117471433B (en) * | 2023-12-28 | 2024-04-02 | 广东威恒输变电工程有限公司 | Construction machinery laser point cloud real-time extraction method based on high reflection intensity target |
| CN117689536B (en) * | 2024-02-01 | 2024-05-10 | 浙江华是科技股份有限公司 | Laser radar splicing registration method, system, device and computer storage medium |
| CN117689536A (en) * | 2024-02-01 | 2024-03-12 | 浙江华是科技股份有限公司 | Laser radar splicing registration method, system, device and computer storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109839624A (en) | A kind of multilasered optical radar position calibration method and device | |
| Pusztai et al. | Accurate calibration of LiDAR-camera systems using ordinary boxes | |
| CN110021046B (en) | External parameter calibration method and system for camera and laser radar combined sensor | |
| CN110689579A (en) | Rapid monocular vision pose measurement method and measurement system based on cooperative target | |
| CN105021124B (en) | A kind of planar part three-dimensional position and normal vector computational methods based on depth map | |
| US20220044411A1 (en) | Detecting crop related row from image | |
| CN111735439B (en) | Map construction method, map construction device and computer-readable storage medium | |
| CN101655982B (en) | Image registration method based on improved Harris angular point | |
| CN104730520B (en) | Circumference SAR back projection self-focusing method based on subaperture synthesis | |
| CN111380510B (en) | Repositioning method and device and robot | |
| CN110599489A (en) | Target space positioning method | |
| CN114152935B (en) | Method, device and equipment for evaluating radar external parameter calibration precision | |
| CN104615880B (en) | Rapid ICP (inductively coupled plasma) method for point cloud matching of three-dimensional laser radar | |
| Pinggera et al. | High-performance long range obstacle detection using stereo vision | |
| CN111123242A (en) | Combined calibration method based on laser radar and camera and computer readable storage medium | |
| CN111179351B (en) | Parameter calibration method and device and processing equipment thereof | |
| CN113763478A (en) | Unmanned vehicle camera calibration method, device, equipment, storage medium and system | |
| CN114972532A (en) | Method, device and equipment for calibrating external parameters between laser radars and storage medium | |
| CN111145264B (en) | Multi-sensor calibration method and device and computing equipment | |
| CN102096920A (en) | Target image-based sub-pixel registering method | |
| CN107797110A (en) | The system imaging method and aperture synthesis projection radiation meter of aperture synthesis projection radiation | |
| Cui et al. | ACLC: Automatic Calibration for non-repetitive scanning LiDAR-Camera system based on point cloud noise optimization | |
| CN112505693A (en) | Interferometric inverse synthetic aperture radar imaging registration method, system and storage medium | |
| CN115100287B (en) | External parameter calibration method and robot | |
| CN114280602A (en) | Near-field three-dimensional imaging system and method for millimeter wave synthetic aperture radiometer |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190604 |
|
| RJ01 | Rejection of invention patent application after publication |